ProcirelHeatsink Thermal Design: Junction Maths Made Simple
Heat removal is an electrical problem in disguise. Temperature difference plays voltage, power plays current, and each interface plays a resistor. Solve the chain from junction to ambient and heatsink selection becomes arithmetic.
> At a glance: 7 minute guide · part 10 of 10 in the amplifiers complete guide track · includes a worked example and a quick-reference table.
## The resistance chain
Tj = Ta + P × (RθJC + RθCS + RθSA). Junction-to-case is fixed by the package; case-to-sink by paste or insulator (~0. 2–1 °C/W); sink-to-ambient by the heatsink itself. You buy the RθSA that lands Tj safely below its limit (125 °C silicon, design to ~85–100 °C).
| Interface | Typical Rθ (°C/W) | Note |
| — | — | — |
| Junction-case (TO-220) | 0.5–5 | Fixed by package |
| Paste joint | 0.1–0.3 | Thin smear |
| Mica + paste | 0.3–0.6 | Electrical isolation |
| Sil-pad | 0.5–1.0 | Clean, insulating |
| Heatsink (small extrusion) | 2–8 | Size and airflow |
## Real-world derating
Datasheet sink ratings assume natural convection in open air. Enclosures, dust and adjacent heat sources degrade them 30–50 % margin is standard practice. Forced air roughly doubles capacity but adds failure modes and noise.
## Assembly details that dominate
Paste quantity (a thin smear), torque. Insulator choice shift case-to-sink resistance more than many sink upgrades. A rough, dry or over-torqued joint wastes the finest extrusion ever machined.
## How to apply this in your build
Work through the sequence below each step assumes the previous one passed. For numbers that need calculating, the linked tools at the end of this guide do the arithmetic instantly.
1. Sum the dissipated power in the device
2. Chain the thermal resistances to ambient
3. Size the heatsink for Tj ≤ ~100 °C worst case
4. Assemble with correct paste and torque; verify by measurement
### Worked example
A regulator dissipating 5 W with RθJC 4 °C/W and a 3 °C/W sink. Tj = 25 + 5 × (4 + 0. 3 + 3) = 86 °C safe. The identical setup with a dried-out joint (1. 5 °C/W interface) reaches 100 °C before the room warms. Run the numbers yourself with the PCB Trace Width and the result should agree to within rounding.
> Practical note from the bench. Our build sheets list every hot part with its chain maths no heatsink gets chosen by anecdote.
## Common mistakes to avoid
– Designing to the 125 °C absolute instead of ~100 °C practical
– Enclosing a sink rated for free convection
– Insulating tabs thermally but not electrically (or vice versa)
## Key takeaways
– The resistance chain the foundation of this guide; revisit it if any measurement here surprises you.
– Real-world derating the foundation of this guide; revisit it if any measurement here surprises you.
– Assembly details that dominate the foundation of this guide; revisit it if any measurement here surprises you.
## Prerequisites and preparation
Before starting: sum the dissipated power in the device and chain the thermal resistances to ambient. Keep the [PCB Trace Width](/tools/pcb-trace-width) and [Electrical Power Calculator](/tools/electrical-power) open every number in the worked example is reproducible. Total time including the bench steps: about 6–7 minutes.
## Who benefits most
Hobbyists meeting this topic for the first time, students who want the version with real numbers instead of abstract symbols. Returning engineers refreshing a corner of the craft. The mistake list alone justifies the visit every entry in it was learned the expensive way.
### Quick reference card
| Aspect | Where to find it in this guide |
| — | — |
| Core theory | The resistance chain |
| Application steps | How to apply this in your build |
| Worked numbers | Worked example |
| Failure modes | Common mistakes to avoid |
## How this fits the amplifiers complete guide track
This guide is one stop in the structured learning path. Start from the [amplifiers complete guide](/tutorial/amplifiers-complete-guide) pillar page for the full map, or continue with [AB output stage heat](/tutorial/class-ab-push-pull-output) and [regulator heat budgets](/tutorial/linear-vs-switching-regulators). For the arithmetic, open the [PCB Trace Width](/tools/pcb-trace-width) or [Electrical Power Calculator](/tools/electrical-power).
## Frequently asked questions
Is bigger always better?
Thermally yes, until mechanical and cost realities object; the maths finds the minimum, taste finds the margin.
Paste or pad?
Paste for lowest resistance, pads for clean repeatability and isolation both beat dry joints.
Is there a calculator for this?
Yes the [PCB Trace Width](/tools/pcb-trace-width) and [Electrical Power Calculator](/tools/electrical-power) tools run the formulas from this guide instantly, client-side, with no signup.
## Keep going with this track
– The complete amplifiers & audio guide: [Amplifiers & Audio complete guide](/tutorial/amplifiers-complete-guide)
– Read next: [inverting op-amp: gain, virtual ground and design](/tutorial/inverting-op-amp-configuration)
– Also in this track: [non-inverting op-amp: high-impedance gain stage](/tutorial/non-inverting-op-amp-configuration)
– Continue with: [the op-amp voltage follower: small circuit, big leverage](/tutorial/op-amp-voltage-follower)
– Calculate as you go: [op-amp gain calculator](/tools/op-amp-gain) · [power dissipation tool](/tools/electrical-power) · [gain to dB converter](/tools/opamp-gain)
– Bookmark this page against the day a measurement surprises you. Most readers return to the table and the mistake list first, and that is the correct order.
## Working method notes
Keep a lab notebook entry for every build in this track. The measured values, the deviations from the guide and the reason for each. Six months from now, those notes are worth more than any tutorial. They describe your bench and your components rather than a general case.
When a result here disagrees with your expectation, write down both numbers before changing anything. The gap between predicted and measured is where the real engineering lives. It is usually a tolerance, a parasitic or an assumption that was never checked.
## Formulas and checks from this guide
Verification checklist for this track: measure bias at idle before signal testing, check heatsink temperature under load. Confirm gain across frequency rather than at 1 kHz alone. Amplifiers forgive nothing at their edges.
Bookmark this page against your next build in the track. The checklist above is the same one used across 19 guides in this series.
## What the bench taught us
Across the emitter resistors at idle, no signal. Compare each output device.
Thermal drift in bias. Check the bias tracker is bolted to the heatsink and re-set per the service values.
## A parting thought for builders
A note on thermal reality, which ends most arguments in this track: dissipation is set at idle, verified under load, and never forgiven by a bigger heatsink alone.
Working through The resistance chainand Real-world derating with that habit in mind takes minutes, and it is the difference between reading about this topic and owning it.